This paper proves that Random Reshuffling in Shuffling Stochastic Gradient Descent dominates vanilla SGD in smooth convex optimization after any finite number of epochs.

This paper proves the conjecture that Local SGD outperforms Mini-batch SGD under bounded second-order heterogeneity for general convex objectives, improving the convergence guarantee and lower bounds.

This paper corrects the theoretical analysis of DP-SGD by identifying that common implementations, which use batch averaging, result in weaker privacy guarantees than previously reported.

The paper introduces Fractional-Order Differentially Private Stochastic Gradient Descent (FO-DP-SGD), a mechanism that incorporates fractional memory into the gradient release process to improve priva…

This paper introduces Curvature-Weighted Gradient Diversity (CWGD), a geometry-aware measure for optimization noise that reduces the asymptotic optimization error floor by up to a factor of two compar…

The paper introduces SMA-DP-SGD, a Spectral Memory-Aware Differential Privacy method that enhances standard DP-SGD by incorporating a memory branch derived from past noisy updates, improving model uti…

The paper introduces Balanced Iteration Subsampling (BIS), a structured sampling scheme that is proven to achieve stronger privacy amplification than the standard Poisson subsampling used in DP-SGD by…

The paper provides a tight, transparent, and closed-form analysis of the trade-off function for Differentially Private SGD using random shuffling, significantly improving upon previous methods and est…

This paper provides the first non-vacuous generalization analysis for the Stochastic Variance Reduced Gradient (SVRG) method by establishing sharp, data-dependent algorithmic stability bounds, thereby…

This paper provides a comprehensive generalization analysis of Stochastic Gradient Descent with Momentum (SGDM) by establishing tight, on-average model stability bounds that show SGDM can generalize w…

The paper proposes an Android-based middleware that enables visually impaired users to securely and independently perform mobile money transactions via voice commands, significantly improving accessib…

This paper empirically evaluates the performance of the Polars DataFrame engine running within Intel SGX2 enclaves, finding that while the overall security overhead is manageable, the performance is s…

The paper proposes PCDM, a diffusion-based framework that enables highly stealthy and effective data poisoning attacks against Federated Learning systems, significantly degrading global performance wh…

The paper proposes Skill-Conditioned Gated Self-Distillation (SGSD), a novel framework that uses retrieved, potentially noisy skills to guide LLM reasoning, achieving state-of-the-art performance on m…

Introduce Deep Second-Order Stochastic Residual Method (D2SRM) for high-dimensional, Hessian-dependent fully nonlinear parabolic PDEs, establish well-posedness, and develop population-level convergenc…

The paper introduces Security Level 5 (SL5), a new, highly stringent security standard for AI systems designed to withstand attacks from state-level, top-tier cyber adversaries.

This paper presents a deterministic algorithm achieving an expected competitive ratio of O(1) for Euclidean online TSP in high dimensions and O(log n) for d = 1, improving upon previous O(sqrt(n)) and…

This paper provides the first systematic threat analysis of State-Space Models (SSMs) in safety-critical applications, introducing novel attack classes and formal metrics to quantify their security an…